Algorithms for artificial intelligence, decision support and therapy, what information should patients provide? How satisfied is it?

Marguerite Brac de la Pierre, Monday 30 May 2022

After addressing the last two parts (1), the obligations of healthcare professionals and the context of prevention, diagnosis or care using artificial intelligence (AI) tools of healthcare organizations, respectively, this is a question in this paper. To specify the conditions under which rights and patients must be satisfied.

As is clearly explained in the last sections, if there is a key word related to artificial intelligence and it may seem as retrograde, it is human, which must be at the center of every stage of design, development and installation. Artificial intelligence system, which is confirmed by legal framework and soft law.

About a year ago, the WHO defined six principles that, in all countries, AI works in the public interest in health, especially human autonomy and transparency.

“AI is a tool that should serve the people and ultimately be a positive force for the society to enhance the welfare of the people. These are justifications and objectives of a proposal for a control over artificial intelligence, which is fully consistent with the Commission’s overall digital strategy to contribute to the promotion of technology in human service. The proposal thus establishes a coherent, functional and proportional framework so that AI develops in a way that respects the rights of the people and earns their trust.

Transparency obligations thus enable individuals to exercise effective remedies and the right to transparency.

This is in the sense that European policies for digital health ethics were adopted by France as President of the Council of the European Union, especially in Article 4.M : “When an artificial intelligence is applied, the maximum is done so that it is interpretable and without discriminatory bias”.

Similarly, the provisions of the Bioethics Act, transferred to the Public Health Code, establish the obligation for health professionals to inform the patient about:

– use, “for prevention, diagnosis or care, a medical device consisting of an algorithmic data processing whose learning has been performed from extensive data”;

– The resulting explanation, which means that the patient must benefit from a light that allows him to appreciate it.

The information provided to patients within the healthcare organization, especially in the Privacy Policy, therefore relies on the services of the AI ​​user, to complement the specific information provided to patients by healthcare professionals regarding the use of an AI system, especially the context of its use, its destination, Its reliability, its effectiveness and its specifications related to the data used.

Ultimately, the data subject has the right not to be the subject of a decision based solely on automated processing that creates or imposes similar legal effects on it, under GDPR, thus echoing the human controls contained in the draft AI regulations.

Everyone understands that transparency, and therefore information, is the main obligation of designers in the case of users and users in the case of patients.

It should be noted that the penalties for data protection and artificial intelligence are the same, however, the latter can amount up to € 20,000,000 in terms of transparency requirements.


(1) https://www.dsih.fr/article/4703/intelligence-artificielle-algorithms-d-aide-a-la-decision-et-a-la-therapeutique-comment-la-garantie-humane-doit -This-bay-implemented-key-needs-for-professional-health-user.html

https://www.dsih.fr/article/4726/intelligence-artificielle-algorithms-d-aide-a-la-decision-et-a-la-therapeutique-quelles-requirements-pour-les-users-des- health-facilities.html


Author

Marguerite Brac de La Perrière is a lawyer, partner at Lerins, Digital Health Specialist.
It helps healthcare players in their regulatory compliance, development and growth, especially in the areas of data processing, data recycling and IT contracts.
mbracdelaperriere@lerinsbcw.com

# Therapeutic#dsih# Patient# Digital# GDPR


Leave a Comment